Nerve cell losses constitute the pivotal event in all deteriorative CNS conditions, whether in degenerative diseases of the abiotrophic, systematic atrophy type, whether in a variety of presenile dementias, or whether ultimately in aging. Yet, good ultrastructural descriptions of specific ways in which the nerve cell is lost; adequate renditions of the changes in the neuropil; reliable accounts of the effects on neighboring or remote but synaptically connected nerve cells; all these descriptions are but meager. Even more, the neurochemical events which accompany degeneration of specific nerve cells and the consequent glial response have barely been touched. To be considered are the changes in (putative) neurotransmitters and amino acid content in regions deprived of specific nerve-ending populations as well as the cyclic nucleotides alterations accompanying the loss of specific neuronal populations. The cerebellum of the Purkinje cell degeneration (pcd) mutant mouse lends itself to such a detailed study. There is a unique homology between the process of nerve cell loss in this model and that seen in many deteriorative CNS conditions in man in that the nerve cells at first seem to develop and mature normally, (at least in the first two weeks of life) establishing a full set of connections, and only then proceed to degenerate. This is the cardinal difference between this particular mutant model and numerous other mutants reported hitherto, with more precocious cerebellar degenerations. Let us summarize, then why the pcd mutant seems to afford such a beautiful model to study neuronal loss and remodeling as well as the concomitant neurochemical events in the CNS. 1. The Purkinje cells seem to develop at first quite normally; 2. then all the Purkinje cells become involved in a relatively short period of time; 3. the architecture of the cerebellar cortex is well known; 4. the neurochemistry of the cerebellum is progessively better understood.